Hunter Fraser | Harrison Truong
Stanford Report News - December 17th, 2025 - Joy Leighton
Natural selection may have favored changes in some of the genes responsible for autism spectrum disorders, according to a study published in Molecular Biology and Evolution.
The results do not necessarily mean that there is an evolutionary advantage to autism itself, said Hunter Fraser, the senior author on the new paper, a professor of biology in the Stanford School of Humanities and Sciences, and an affiliate of the Wu Tsai Neurosciences Institute. Instead, the evidence suggests that whatever it is that led to the evolution of powerful, complex human brains may have led as well to autism and other neurological disorders, including schizophrenia.
Autism and schizophrenia were not, however, what Fraser and his graduate student Alex Starr set out to understand. Originally, the pair wanted to test a broader hypothesis about how quickly different kinds of brain cells evolved over the millennia.
“We know that there are many different types of neurons in the brain and that they evolved at different rates,” Starr said, “but it isn’t well understood why.”
To tackle that issue, Starr and Fraser took inspiration from a fact about protein evolution. In general, proteins that are more abundant in the body tend to evolve more slowly than others. Scientists believe this slow pace is protective: Detrimental mutations in common proteins could be disastrous for an organism. In contrast, a harmful mutation in a less ubiquitous protein might not be great but would be less likely to kill an animal right away. As a result, mutations in less common proteins are generally better tolerated by natural selection.
Starr and Fraser figured something similar might be true of cells, so they set out to see whether abundant cell types might have evolved more slowly than rarer ones. For this study, they focused on cells in the neocortex, the outer layer of the mammalian brain that plays key roles in human language, sensory perception, and other core mental operations. Then, they scoured existing datasets for information on gene expression in individual neocortical cells, looking to see how much variation there was across different animals.
“If you see that gene expression is almost identical between similar cell types in mice and humans, for example, that’s evidence that evolution has been slow, and if it’s very different, evolution has been acting more quickly,” Starr said.
As they expected, Starr and Fraser found that the most common cell types in the neocortex tended to vary less across different animals – evidence that those cell types evolved more slowly than less common ones.
Across primates, there were very few exceptions. However, there was a notable one specific to human evolution: a cell type called layer 2/3 intratelencephalic excitatory neurons, or L2/3 IT neurons, named because of its location in the second and third layers of the six-layer cortex. Those neurons play a critical role in enabling communication between different parts of the neocortex, a key component of advanced cognition. But despite those neurons’ abundance, the pattern of gene expression that orchestrates their inner workings showed signs of having evolved quite rapidly during human evolution but not in the evolution of other mammals.
At the same time, around 100 genes thought to protect against autism were expressed less in human 2/3 IT neurons compared with the same cells in chimpanzees. Much the same was true of genes that protect against schizophrenia, the researchers found.
Those observations hinted that evolutionary pressures sped up changes to gene expression in 2/3 IT neurons. That is, it appeared that something about the way those neurons changed gave humans an evolutionary advantage, a conclusion the researchers backed up with additional genetic and statistical tests, including studying human and chimpanzee brain organoids in the lab.
Unfortunately, it’s hard to say what exactly the advantage was, Starr said. One possibility relates to the fact that human brains take longer to grow and develop compared with other primates, including chimpanzees. Biologists believe that relatively slow growth made possible our more complex, flexible, and computationally powerful brains. It’s possible, Starr said, that the same genes that slowed our brains’ growth also resulted in an increased risk of autism and schizophrenia – although he cautions there is not yet any evidence to support or refute the hypothesis.
Whatever the exact reasons, the results indicate that autism could be tied directly to how our brains evolved. “Neurodiversity could be an essential part of being human,” Starr said.
Fraser is also a member of Bio-X, the Maternal & Child Health Research Institute, and the Stanford Cancer Institute.
This story was originally published by Stanford School of Humanities and Sciences.
